UBC Faculty Research and Publications

The Effects of Three Worksite Wellness Interventions to Promote Fruit and Vegetable Consumption and Weight… Gotay, Carolyn C., 1951-; Monro, Melody; Shen, Hui; Amick, Benjamin C.; Bottorff, J. L.; Corbett, Kitty K.; MacPhail, Sue; Storoschuk, Sharon Aug 26, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-Gotay_C_et_al_Effects_Three_Worksite.pdf [ 1.92MB ]
JSON: 52383-1.0372021.json
JSON-LD: 52383-1.0372021-ld.json
RDF/XML (Pretty): 52383-1.0372021-rdf.xml
RDF/JSON: 52383-1.0372021-rdf.json
Turtle: 52383-1.0372021-turtle.txt
N-Triples: 52383-1.0372021-rdf-ntriples.txt
Original Record: 52383-1.0372021-source.json
Full Text

Full Text

  1   The Effects of Three Worksite Wellness Interventions to Promote Fruit and Vegetable Consumption and Weight Loss   Carolyn C Gotay1, PhD, Melody Monro1, MPA, Hui Shen1, PhD, Benjamin C. Amick III2, PhD, Joan L. Bottorff3, PhD, Kitty K. Corbett4, PhD, Sue MacPhail5, and Sharon Storoschuk5, MPH    1 School of Population and Public Health, University of British Columbia, Vancouver, BC 2 Robert Stempel College of Public Health and Social Work, Florida International University, Miami, Florida, USA; Institute for Work & Health, Toronto, ON  3 Institute for Healthy Living and Chronic Disease Prevention, University of British Columbia, Kelowna, BC; Faculty of Health Sciences, Australian Catholic University, Melbourne, Victoria  4 School of Public Health and Health Systems, University of Waterloo, Waterloo, ON 5 Independent Contractor, Vancouver, BC  Corresponding author: Carolyn C Gotay, PhD, School of Population and Public Health, University of British Columbia, 2206 East Mall, Vancouver, BC Canada V6T 1Z3. email: carolyn.gotay@ubc.ca   Source of Funding: Funding for this project was provided by the Canadian Cancer Society Grant #021032  Conflicts of Interest: None to declare  Trial registry name:  Clinical Trials.gov Registration number NCT02535754 Date of registration August 26, 2015        2        ACKNOWLEDGEMENTS We are grateful for the assistance of staff and participation of employees at the University of British Columbia, Okanagan campus (Kelowna, BC), University of the Fraser Valley (Abbotsford, BC), and Thompson Rivers University (Kamloops, BC). We appreciate the input of Barbara Sternfeld, Glorian Sorensen, NutritionQuest, and Marliese Dawson.              3 ABSTRACT Objective: This study investigated the impact of three different worksite approaches to healthy behavior change: a personalized individual intervention; a comprehensive program using environmental and social support; and both approaches combined.  Methods: 680 individuals at three educational institutions participated in a year-long intervention. The primary outcome was change in fruit and vegetable intake from baseline to four months post-intervention completion tested by linear mixed effect (LME) models. Results: Significant increases in fruit and vegetable consumption were seen in the individual and combined conditions, with the greatest increase in the individual condition. Conclusions: The superiority of the individual intervention implies that for well-defined and concrete outcomes, a clear, consistent, and frequently repeated message has the most impact.          4 INTRODUCTION Working Canadians spend on average almost half of their waking hours - more than 36 hours from Monday to Friday - on work activities,1 and Americans spend even more: 38.6 hours per week.2 As such, work sites represent a prime target for programs aimed at health promotion and disease prevention. Workplace wellness programs (WWPs) provide a platform for broad participation, especially when initiatives take place during the workday, are provided on-site, and involve stable populations of workers.3 They also have the benefits of building on intact social networks, in-place communication systems and events, and existing facilities and professional resources.  Considerable data support the effectiveness of WWPs: as O’Donnell4 states, “Hundreds of methodologically sound studies have shown that well-designed workplace health promotion programs are effective in improving health. ” A comprehensive analysis by Sorensen and Quintiliani5 concluded that “research has documented the efficacy of these programs across a wide variety of outcomes, including changes in anthropometric measures, health behaviors, life satisfaction indicators, and measures of morbidity and mortality” (p. 260). With the increase in preventable chronic diseases across the world, health promotion and disease prevention have become increasingly important global concerns to employers and employees alike. In particular, improving eating patterns and reducing obesity are areas where priority attention is warranted, given the rapid increases in obesity rates, with implications for heart disease, diabetes, a number of cancers, stroke, and many other diseases.  Research on WWPs has shown promising results in improving nutrition6-8 and decreasing obesity.9 The most common approaches target individual behavior change through education and counselling.9 While such programs can be successful, they may not be sustainable at the   5 conclusion of the research project. New technologies offer novel opportunities to reach workers in sophisticated ways at a relatively low cost that may foster sustainability. Allen. et al. ’s systematic review of the literature from 2002-1210 concluded that “technology-assisted interventions” for weight maintenance or loss were promising. Such interventions in  worksites have found positive effects on nutrition, including fruit and vegetable consumption, using tailored emails11 and on weight loss using text messages12 and internet-delivered programming.13    There are fewer examples of successful nutrition and weight control approaches that target changes in the work environment, such as changing vending machine content and providing nutritional information in cafeterias. Two large and rigorous trials found that worksite interventions that primarily emphasized environmental change did not result in weight loss.14,15 Fewer still worksite interventions have reported on the impact of interventions based on social interaction among employees. One such program, which also included environmental changes, used team challenges and walking clubs, among other strategies based on increasing social support for obesity prevention.16  This study found that the intervention was not successful in leading to obesity reduction overall, although a sensitivity analysis showed that individuals with more participation in the social activities experienced more benefits.  We have not found studies that explore the relative effectiveness of approaches directed at individuals vs. the social and physical environments, or the impact of combining the two approaches compared to either one alone. The purpose of this study was to investigate the impact of three different worksite approaches to healthy behavior change: a personalized technology-driven individual intervention; a comprehensive program using environmental and   6 social support; and both approaches combined. The primary outcome was change in fruit and vegetable intake from baseline to four months post-intervention completion (16 months post-accrual) and weight loss as a secondary outcome.  METHODS Study design  The study used a community intervention design in which three institutions were randomly assigned to one of three year-long intervention conditions. Assessments were taken at baseline and 16 months later. Data analysis was based on linear mixed effect model.  Setting/participants  A number of worksites in British Columbia (BC), Canada were approached for possible study participation. Institutional requirements were availability of individual employee email addresses that could be accessed by the research team for study procedures, agreement of a company employee to serve as a study contact, and access to locations to hold worksite events as part of the project. Three institutions that met study requirements were invited and agreed to take part in the study. All are public universities located outside main BC urban centres. This study was reviewed and approved by the University of British Columbia Behavioural Research Ethics Board (H10-01079). Study participation was completely voluntary, all participants were fully informed and gave consent, and all responses were kept anonymous.  The universities were randomly assigned to the following study conditions: (1) an empirically-validated intervention directed at individual behavior change using personally-tailored messages delivered by email (ALIVE: A Lifestyle Intervention Via Email);11 (2) a comprehensive approach building on social and institutional support developed by the Canadian   7 Cancer Society British Columbia Yukon and previously used for worksite tobacco control (CCS) and (3) an intervention including both of these approaches (combined).  Our analysis plan was powered on at least 80 participants in each condition, with significance level 0.05 and a power of 90% to detect a medium effect size under the assumption of intra-cluster correlation coefficient (ICC) 0.002.17  After randomization of the institution to condition, participants were recruited during a special promotional event held at each campus –to explain the purpose of the study, describe the intervention, introduce study personnel, identify and establish relationships with “workplace champions,” and complete baseline questionnaires with eligible participants. Eligibility criteria included being an employee of the company (not a visitor or student); English speaking with reading level at minimum Flesch-Kincaid 8.5 grade level; and  ability and willingness to provide access to individual work email address. Baseline questionnaire completion was a requirement to enrol in the ALIVE conditions (as the questionnaire responses were used to generate the personally-tailored email messages). Completing the questionnaire was encouraged in the CCS group, either at the kick off event or in the next few weeks afterwards. Data were collected 2000-212 and analyzed 2012-16.  Interventions  ALIVE condition. The ALIVE program was shown to be efficacious in a previous randomized controlled trial (RCT) in a worksite.11,18 The program is based on theories of health behavior change and provides computer-generated, individually tailored messages delivered by email that include reinforcement, goal-setting and tips to achieve these goals, interactive features, and   8 health information. Information used to tailor the intervention is obtained through the baseline questionnaire.  We used a revised ALIVE intervention in this study that had just been finalized by the company that manages this program (NutritionQuest).19 The materials were also modified slightly to accommodate Canadian spelling and verbal nuances. The intervention we used retained all of the features of the original program but  with a more sustained focus on behavior change in a particular area, either diet (fruit and vegetable consumption) or physical activity. Participants selected their preferred area and received a 24 week program related to that risk factor; they then changed to the other risk behavior. The initial protocol called for twice weekly emails to be sent for throughout the total intervention period of 48 weeks. Due to participant demand, we reduced the frequency of emails to once per week midway through the intervention period.  CCS condition. This condition provided a year-long program focusing on nutrition and physical activity, led by the CCS and modeled after the Society’s previous successful tobacco control worksite program. The program stresses the importance of developing social and environmental supports for behavior change within the worksite. The specific supports varied according to the constraints and priorities of the individual worksite, but in both companies that received this intervention, activities focused on modifying food availability (e.g., cafeteria and vending machine selections); social environment (e.g., group lunchtime walking clubs), special events (e.g., healthy cooking classes, zumba instruction) and contests (e.g., units competing to accrue the most steps). The CCS provided resources, information, tools, and other support as needed and worked in coordination with onsite staff.    9 Combined condition. In this arm, participants received both ALIVE and the CCS programs.  Main outcome measures  The primary outcome was change in self-reported number of daily servings of fruits and vegetables from baseline to 16 months, as an indicator of healthy diet. Change in weight was a secondary outcome. We also assessed self-efficacy for behavior change, health perceptions, work performance, and participant evaluation of the program. We intended to assess physical activity, but this was not possible due to a programming error. Findings on the other measures will be reported in additional publications.  Statistical analysis Descriptive statistics for demographic variables were summarized for each intervention site. Continuous variables were presented by mean and standard deviation and categorical variables were displayed by frequency and percentage.  Statistics for primary outcomes were summarized at baseline and final phase by intervention site. Linear mixed effect (LME) models were used to compare the outcome difference between the baseline and final phase within each intervention and between the three interventions. When LME models were fit, gender and age were controlled for the within group analysis, and baseline values were further controlled for the between group analysis. LME was implemented in this study since the method has advantages in dealing with intra-cluster correlation and missing data.  Analyses were conducted using SAS 9 4. All statistical tests were two-sided, with a significance level of 0.05.     10 RESULTS Participant flow is shown in Figure 1. A total of 680 individuals completed baseline questionnaires: 170 in the ALIVE condition, 285 in the CCS condition, and 225 in the combined condition. There was considerable attrition in completion of the final questionnaire, even though we added an incentive (coffee cards) when we determined that completion was lagging. Completion rates were 72 (42.4%) in the ALIVE condition, 179 (62.8%) in the CCS condition, and 109 (48.4%) in the combined condition. As described below, we examined the impact of non-completion on the results. For the weight outcome, we excluded individuals who reported that they had not weighed themselves recently (“in the past few weeks”) in either the baseline or final questionnaire, which yielded n= 120 in the Alive condition, 191 in the CCS condition, and 169 in the combined condition at the baseline. Results were very similar when all weight responses were included.  Table 1 summarizes baseline descriptive data. The participants were mostly middle-aged, primarily women, white, and non-smokers. Around one-third reported one or more chronic diseases (from a list of 8 including high blood pressure and diabetes). There were no significant differences across conditions except for age, where participants in the ALIVE condition were younger than those in the other two groups (p=0.0001).  Table 2 provides statistics for the primary outcome - number of servings of fruits and vegetables eaten yesterday - and the secondary outcome - weight - at baseline and final phase, by intervention site. At baseline, individuals reported eating just over three servings per day in all sites, and there were no significant differences across sites (p=0.9). It should be noted that BMI calculations based on self-reported baseline weight and height were 26.1, 26.5, and 26.9   11 indicating the average respondent was overweight in all three sites, with no significant difference across sites (p=0.4).  The post-intervention data showed statistically significant increases in fruit and vegetable consumption in Alive and combined conditions and decreases in weight from baseline to final phase in Alive and CCS conditions. There was increased  fruit and vegetable consumption in the CCS condition and decreased weight in the combined condition but these differences were not statistically significant. The increase in mean fruit and vegetable consumption between the baseline and final scores ranged from 0.24 servings to 1.12 servings per day. ALIVE showed the biggest increase with more than one serving per day. The decrease in mean weight ranged from 1.78 lbs to 4.38 lbs. ALIVE demonstrated the biggest weight loss.  Given the considerable non-completion of the final questionnaire, we examined potential bias by comparing individuals who completed both questionnaires vs. those who did not on baseline descriptors: age, gender, weight, BMI, and fruit and vegetable consumption. We conducted separate analyses for each site since completion rates varied across sites. There was only one significant difference: women were more likely than men to complete the final questionnaire in the ALIVE condition (p=0.01).  Given this difference (albeit small), we decided to control for gender and age (given that participants were younger in one site) for within intervention analysis.  Table 3 shows that from baseline to final assessment, ALIVE generated a greater fruit and vegetable consumption and less weight than the CCS or combined conditions after controlling for gender, age, and baseline value. The between-intervention difference was statistically significant for fruit and vegetable consumption but not for weigh. The comparisons among the   12 three interventions indicate that ALIVE was the most effective for improving fruit and vegetable consumption and controlling weight while there was not a significant difference between CCS and combined conditions.  CONCLUSIONS This report describes the results of a community intervention trial of three approaches to worksite health promotion that varied along the dimension of individual vs. group interventions. The interventions were implemented in real world settings where the program activities took place in the context of many other events in the environment – such as the key contact in one of the institutions taking another position shortly after the study began. The successful completion of a year-long intervention in three different locations, each with different constraints, was due to consistent oversight of the research team, particularly the study coordinator (MM), and the committed staff at the worksites. The findings were rewarding: the data showed that all three interventions had a positive impact on fruit and vegetable intake and weight reduction, but that the individually-focused approach – in which regular email messages were tailored to the individual based on his or her own preferences, knowledge, and attitudes (such as self-efficacy and awareness) led to the best results for both outcomes. With respect to well-defined and concrete outcomes, such as increasing fruit and vegetable consumption or losing weight, it may be that a clear, consistent, and frequently repeated message has the most impact. Our research team has labeled this “The Mom Effect,” since it reminded us of how our mothers shaped our behaviors when we were growing up.  We were surprised that the combined intervention – which provided both the individual tailoring and group-based programming – did not result in the best outcomes, since it provided   13 participants with more options for change. It is not clear why this intervention did not stand above the others, except to note that while the sites had major similarities, they also varied in many ways. Although there were virtually no differences between participants in different sites on baseline measures, it may be that other factors we did not measure affected the effectiveness of the combined intervention. It is also possible that sometimes interventions that intervene in multiple ways – through email and ongoing social activities and environmental changes – may be overwhelming to the point that workers turn off all the messages. More is not necessarily better.  We learned this lesson throughout the study. One study weakness is the loss to follow up of a considerable number of participants. The statistical analyses took this into account to the degree possible – through conducting linear mixed effect models – and our investigation of baseline differences on key descriptive and behavioral variables did not reveal differences between completers and non-completers. However, it is possible that there were differences between those who completed the baseline and follow-up questionnaires that we did not measure that could limit generalizability. Further, generalizability is limited by the sample size that only reflected a small proportion of the workforce and the small numbers of men who took part in the study. Worksite health promotion programs may need to tailor programs to men’s values and preferences to increase their interest and participation.20-22 A major factor that the participants brought to our attention was the burden posed by frequent emails. At the beginning of the study, email was still a bit of a novelty, and people looked forward to seeing their messages. As the study progressed, the social phenomenon of burdensome emails increased. Participants in the study began to block NutritionQuest (NQ) from sending them emails. As we explored this problem, we became aware that we would have to   14 decrease the number of emails we were sending, and we modified the original ALIVE protocol so that participants received one rather than two emails per week. When this was announced, some participants “unblocked” NQ from their email address and they could continue to receive messages. However, given this scenario, it is perhaps not surprising that completion of the final questionnaire was much higher in the site that didn’t have the emails (62.8%) than in the two sites that did (42.4% and 48.4%). This experience vividly demonstrates that using “new” technology in research is a tricky business, as the attraction of a particular technological advance is fleeting.  Another limitation in the study is its reliance on self-report measures. In natural settings such as the workplaces in this study, we did not have the resources to, for example, bring all the participants in and weigh them at baseline and 16 months later. We requested and received initial commitment to obtain absentee and health records from the companies, but at the conclusion of the study, they were unwilling to provide these to us. If other researchers have such resources available, we strongly support the use of objective measures. While response biases such as social desirability can never be ruled out when it comes to self-reports, we believe that the long lag period between the initial questionnaire and the follow-up survey (16 months) and the modest positive changes in fruit and vegetable consumption and weight mitigate against social desirability bias.  Since we did not have a “no treatment control group,” we cannot rule out changes due to other events happening during this time period; perhaps the entire populations of the universities ate healthier and lost weight during the time of the study. Although we cannot definitively dismiss this hypothesis, population-based data on these outcomes are available from   15 the annual Canadian Community Health Surveys administered during the period the study was being conducted. These data show that for British Columbia adults, the proportions who reported daily consumption of fruits and vegetables went from 42.3% in 2010 down to 42.1% in 2012. For proportions who reported being overweight or obese, the figures were 44.4% in 2010 and 46.5% in 2012.23 These data do not suggest a widespread temporal trend toward positive health behaviors in the province, rather, the opposite.  The positive changes in nutrition and weight were modest. However, in the public health context, such small changes in the population can have major benefits to society. Krueger et al.24 have shown that even a 1% relative risk reduction in risk factors including excess weight, smoking, and physical inactivity have considerable economic and health benefits for the population.  Continued health promotion programs in the worksite such as the interventions used in this study have the potential to contribute to these changes. Future programs need to take into consideration changes in the work landscape (e g., more telecommuting), the labor force (e.g., more older workers) and technology (e.g., more reliance on apps and increased smart phone functionality) to capitalize on this potential.            16 REFERENCES 1.  Statistics Canada. Paid work and related activities. https://www. statcan. gc. ca/pub/89-647-x/2011001/hl-fs-eng. htm. Accessed April 3, 2018.  2.  Organisation for Economic Cooperation and Development. Average annual hours actually worked per worker. https://stats. oecd. org/Index. aspx?DataSetCode=ANHRS. Accessed April 3, 2018.  3.  Harden A, Peersman G, Oliver S, Mauthner M, Oakley A. A systematic review of the effectiveness of health promotion interventions in the workplace. Occup Med (Lond). 1999;49(8):540-548.  4.  O'Donnell M. Does workplace health promotion work or not? Are you sure you really want to know the truth? American journal of health promotion : AJHP. 2013;28(1):iv-vi.  5.  Sorenson G, Quintiliani L. Effective Programs to Promote Worker Health Within Healthy and Safe Workplaces. In: Pronk NP, ed. ACMS's Worksite Health Handbook : A Guide To Building Healthy And Productive Companies                2ed. Champaign, IL. American College of Sports Medicine; 2009:259-368.  6.  Geaney F, Kelly C, Greiner BA, Harrington JM, Perry IJ, Beirne P. The effectiveness of workplace dietary modification interventions: a systematic review. Prev Med. 2013;57(5):438-447.  7.  Maes L, Van Cauwenberghe E, Van Lippevelde W, et al. Effectiveness of workplace interventions in Europe promoting healthy eating: a systematic review. Eur J Public Health. 2012;22(5):677-683.    17 8.  Thomson CA, Ravia J. A systematic review of behavioral interventions to promote intake of fruit and vegetables. J Am Diet Assoc. 2011;111(10):1523-1535.  9.  Anderson LM, Quinn TA, Glanz K, et al. The effectiveness of worksite nutrition and physical activity interventions for controlling employee overweight and obesity: a systematic review. Am J Prev Med. 2009;37(4):340-357.  10.  Allen JK, Stephens J, Patel A. Technology-assisted weight management interventions: systematic review of clinical trials. Telemed J E Health. 2014;20(12):1103-1120.  11.  Sternfeld B, Block C, Quesenberry CP, Jr. , et al. Improving diet and physical activity with ALIVE: a worksite randomized trial. Am J Prev Med. 2009;36(6):475-483.  12.  Kim JY, Oh S, Steinhubl S, et al. Effectiveness of 6 months of tailored text message reminders for obese male participants in a worksite weight loss program: randomized controlled trial. JMIR Mhealth Uhealth. 2015;3(1):e14.  13.  Ross KM, Wing RR. Implementation of an Internet Weight Loss Program in a Worksite Setting. J Obes. 2016;2016:9372515.  14.  Williams AE, Stevens VJ, Albright CL, Nigg CR, Meenan RT, Vogt TM. The results of a 2-year randomized trial of a worksite weight management intervention. American journal of health promotion : AJHP. 2014;28(5):336-339.  15.  LaCaille LJ, Schultz JF, Goei R, et al. Go!: results from a quasi-experimental obesity prevention trial with hospital employees. Bmc Public Health. 2016;16.  16.  Lemon SC, Zapka J, Li WJ, et al. Step Ahead A Worksite Obesity Prevention Trial Among Hospital Employees. American Journal of Preventive Medicine. 2010;38(1):27-38.    18 17.  Hemming K, Girling AJ, Sitch AJ, Marsh J, Lilford RJ. Sample size calculations for cluster randomised controlled trials with a fixed number of clusters. Bmc Med Res Methodol. 2011;11.  18.  Block G, Sternfeld B, Block CH, et al. Development of Alive! (A Lifestyle Intervention Via Email), and Its Effect on Health-related Quality of Life, Presenteeism, and Other Behavioral Outcomes: Randomized Controlled Trial. J Med Internet Res. 2008;10(4).  19.  Nutritionquest. ALIVE. http://nutritionquest. com. Accessed April 3, 2018.  20.  Bottorff JL, Seaton CL, Johnson ST, et al. An Updated Review of Interventions that Include Promotion of Physical Activity for Adult Men. Sports Medicine. 2015;45(6):775-800.  21.  Caperchione CM, Vandelanotte C, Kolt GS, et al. What a Man Wants: Understanding the Challenges and Motivations to Physical Activity Participation and Healthy Eating in Middle-Aged Australian Men. Am J Mens Health. 2012;6(6):453-461.  22.  Taylor PJ, Kolt GS, Vandelanotte C, et al. A review of the nature and effectiveness of nutrition interventions in adult males - a guide for intervention strategies. Int J Behav Nutr Phy. 2013;10.  23.  Canadian Community Health Survey. Lifestyle and social conditions. http://www5. statcan. gc. ca/cansim/a33?RT=TABLE&themeID=2968&spMode=tables&lang=eng. Accessed April 3, 2018.  24.  Krueger H, Turner D, Krueger J, Ready AE. The economic benefits of risk factor reduction in Canada: Tobacco smoking, excess weight and physical inactivity. Can J Public Health. 2014;105(1):E69-E78.    19 TABLE 1. Participant Characteristics at Baseline Descriptors Alive (n=170) CCS (n=285) Combined (n=225) Age    Mean years  (SD) 39.6 (12.8) 45.4 (11.3) 46.2 (9.5) Gender: n (%)    Female 136 (80%) 245 (86.0%) 191 (84.9%) Male 34 (20%) 40 (14.0%) 34 (15.1%) Race: n (%)    White 146 (85.9%) 247 (86.7%) 205 (91.1%) Non-white 24 (14.1%) 38 (13.3%) 20 (8.9%) Chronic Disease: n (%) Yes 53 (31.2%) 101 (35.4%) 70 (31.1%) No 117 (68.8%) 184 (64.6%) 155 (68.9%) Smoking: n (%)    Non-smoker 158 (92.9%) 270 (94.7%) 214 (95.1%) Smoker 12 (7.1%) 15 (5.3%) 11 (4.9%)  Abbreviations:  SD,  standard deviation         20 TABLE 2. Comparison of Outcomes and Changes Between Baseline and Post-intervention Within the Three Interventions, with Significant Results in Bold   Outcome variable  Alive  CCS  Combined Yesterday fruit and vegetable servings count    Baseline mean (sd) 3.29 (1.78) 3.33 (1.58) 3.33 (1.55) post-intervention mean (sd) 4.42 (2.09) 3.56 (1.60) 3.62 (1.48) Estimated change between post-intervention and baseline (95% CI), using LME 1.12 (0.61 – 1.64) P<0.0001 0.24 (-0.02 – 0.49) P=0.07 0.32 (0.009 – 0.63) P=0.04 Weight (lbs)    Baseline mean (sd) 165.48 (36.70) 163.74 (33.86) 165.98 (40.90) post-intervention mean (sd) 158.61 (32.36) 157.73 (29.83) 167.73 (39.76) Estimated change between post-intervention and baseline (95% CI), using LME -4.38 (-7.42 -  -1.33) P=0.006 -2.38 (-4.29 -  -0.48) P=0.01 -1.78 (-4.76 – 1.21) P=0.2 Note:  Missing values were not imputed. Outcome changes between post-intervention and baseline were computed using LME, controlling for age and gender.  Abbreviations:  SD, standard deviation CI, confidence interval. LME, linear mixed effect.         21 TABLE 3. Comparison of Outcome  Between Baseline and Post-intervention Among the Three Interventions, using LME, with Significant Results in Bold  Outcome variable ALIVE vs. CCS (95% CI) ALIVE vs. Combined (95% CI) Combined  vs. CCS (95% CI) Yesterday fruit & vegetable servings  count 0.24 (0.06 – 0.42) P=0.008 0.19 (0.001 - 0.38) P=0.05 0.05 (-0.11 – 0.21) P=0.5 Weight (lbs) -0.52 (-1.65 – 0.61) P=0.4 -0.46 (-1.65 – 0.72) P=0.4 -0.05 (-1.05 – 0.95) P=0.9 Note: Comparison of outcome among the three interventions was based on LME, controlling  for age, gender and baseline value   Abbreviations:   CI, confidence interval   LME, linear mixed effect                         22 FIGURE 1. Participant flow chart                     *Sum of the number at allocation + the number that completed 16-month follow-up Enrollment Allocation ALIVE N=170  Combination N=225 CCS N=285 16-month follow-up  72 (42%) Completed  98 (58%) Lost to follow-up  109 (48%) Completed  116 (52%) Lost to follow-up  179 (63%) Completed  106 (37%) Lost to follow-up  Analysis  N=242* N=334* N=464* 680 employees                              at 3 institutions  Randomized (n=3) (bbbbbbb(nnnnnn(((o0 excluded 23   APPENDIX: Selected Project Photos   Chef David, University of the Fraser Valley  Chef Ricardo, University of BC Okanagan  Chef David Tombs, Thompson Rivers University  Study Participants  Group Exercise at Final Event  L-R: Jerilynn Maki, Carolyn Gotay, Joan Bottorff  24   APPENDIX: Selected Project Photos   UBC Okanagan Final Event  Be Well at Work swag  Be Well at Work, Vancouver Meeting  Melody Monro, Be Well at Work Project Manager  University of the Fraser Valley Final Event  Thompson Rivers University Final Event  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items